\bigskip         \subcaption*{\normalsize Panel B: Clustered Standard Errors }         \vspace{-0.5ex}         \setlength{\tabcolsep}{6pt}         \begin{tabularx}{0.8\hsize}{lCCCC}         \midrule
                    &\multicolumn{1}{c}{(1)}         &\multicolumn{1}{c}{(2)}         &\multicolumn{1}{c}{(3)}         &\multicolumn{1}{c}{(4)}         \\
\midrule
HEE                 &       0.083\sym{***}&       0.083\sym{***}&       0.092\sym{***}&       0.065\sym{**} \\
                    &     (0.011)         &     (0.019)         &     (0.018)         &     (0.028)         \\
HEEC                &       0.093\sym{***}&       0.075\sym{***}&       0.116\sym{***}&       0.067\sym{*}  \\
                    &     (0.014)         &     (0.020)         &     (0.023)         &     (0.033)         \\
\midrule
\(F\)-test, HEE = HEEC&                     &                     &                     &                     \\
$\qquad$ \(F\)-statistic&       0.757         &       0.151         &       1.497         &       0.005         \\
$\qquad$ \(p\)-value&       0.388         &       0.699         &       0.226         &       0.941         \\
Control Mean        &       0.530         &       0.392         &       0.603         &       0.521         \\
Control SD          &       0.157         &       0.267         &       0.243         &       0.351         \\
N                   &         755         &         755         &         755         &         755         \\
\midrule
\end{tabularx}
\captionsetup{justification=justified,
width=0.8\linewidth}
\caption*{\footnotesize
\textit{Notes:}
This
table
examines
the
robustness
of
the
results
presented
in
Table
\ref{table-midline-knowledge-aggregate}
to
different
regression
specifications.
Panel
A
excludes
randomization
strata
fixed
effects
from
the
regression.
Panel
B
uses
standard
errors
clustered
at
the
wave-class
level.
***$\,p
<
0.01$,
**$\,p
<
0.05$,
*$\,p<0.10$.
}
\end{table}
